TRACE: Transformer-based Risk Assessment for Clinical Evaluation

TL;DR

TRACE introduces a Transformer-based approach for clinical risk assessment that effectively integrates diverse data types, handles missing data, and provides interpretable results to support clinical decision-making.

Contribution

The paper presents a novel Transformer-based architecture for clinical risk assessment that outperforms traditional methods and offers enhanced interpretability.

Findings

Outperforms baseline models in risk detection accuracy

Handles multiple data modalities and missing values effectively

Provides interpretable results via attention weights

Abstract

We present TRACE (Transformer-based Risk Assessment for Clinical Evaluation), a novel method for clinical risk assessment based on clinical data, leveraging the self-attention mechanism for enhanced feature interaction and result interpretation. Our approach is able to handle different data modalities, including continuous, categorical and multiple-choice (checkbox) attributes. The proposed architecture features a shared representation of the clinical data obtained by integrating specialized embeddings of each data modality, enabling the detection of high-risk individuals using Transformer encoder layers. To assess the effectiveness of the proposed method, a strong baseline based on non-negative multi-layer perceptrons (MLPs) is introduced. The proposed method outperforms various baselines widely used in the domain of clinical risk assessment, while effectively handling missing values. In terms of explainability, our Transformer-based method offers easily interpretable results via attention weights, further enhancing the clinicians' decision-making process.